Background/Purpose: While changes in disease severity, functional status, and patient wellbeing are commonly measured in AS trials, longitudinal data from real-world cohorts are fewer; particularly considering the impact of anti-tumor necrosis factor (TNF) therapy. The objective was to estimate changes in severity, function, and wellbeing over time, from a population of Canadians undergoing active management for AS.

Methods: This real-world analysis used up to 15 years of clinical and patient-reported outcomes (PRO) data from the broad set of patients treated at the IRRM and CORQ (RHUMADATA®). The frequency of use of anti-TNFs was tabulated. AS severity (by BASDAI), functional impact (by BASFI), and wellbeing (by BAS-G) were assessed; at the baseline visit, within the first year of follow-up (e.g. from second visit to 52 weeks), and at least annually thereafter. From those with PRO data post-baseline, mean (standard deviation [SD]) scores were plotted over the next 4 years; stratified by anti-TNF status (treatment-naïve; or treated with 1, 2, or ≥3 anti-TNFs over the period). A sensitivity analysis was performed to limit the cohort to those with ≥2 PRO measures post-baseline.

Results: Mean (SD) age at baseline was 36.0 (12.6) years, and 60.0% were male; median follow-up was 9.7 (10.6) years. Of the 341 patients with post-baseline PRO measures, 116 (34.0%) were never treated, 144 patients (42.2%) were treated with 1, 47 patients (13.8%) were treated with 2, and 34 (10.0%) patients were treated with ≥3, anti-TNFs. Mean (SD) baseline PRO scores were higher with greater anti-TNF treatment (table); for example, BASDAI scores ranged from 4.1 (2.2; anti-TNF-naïve), to 5.5 (1.9; ≥3 anti-TNFs). Mean (SD) scores decreased over time, in each anti-TNF treatment category. As an example for the BASDAI, scores decreased to 3.0 (2.1) during year four (anti-TNF-naïve), and to 4.0 (1.7) for those with ≥3 anti-TNFs. Mean changes from baseline were less among those with more severe disease who received more anti-TNF treatment (data not shown). While the sample for the sensitivity analysis was smaller, the findings were consistent with the base case: iimprovements on the BASDAI, BASFI, and BAS-G were lower for patients exposed to ≥2 anti-TNFs, compared to those treated with 0 or 1 (data not shown).

Conclusion: The need for re-treatment was common in this Canadian cohort; almost 25% were treated with ≥2 anti-TNFs over 4 years. PROs showed improvements in disease severity, function, and patient wellbeing over time; although consistent with observations from rheumatoid arthritis, these were less pronounced for more severely affected AS patients (treated more heavily with anti-TNFs). These data are useful for demonstrating the clinical and functional burden experienced by AS patients, and suggest a potential role for newer treatments not targeting TNF.

Table: Mean (SD) scores over time, BASDAI, BASFI, and BAS-G; according to number of anti-TNFs (naïve, 1, 2, or ≥3 therapies)